1. Technical Field
The present invention relates to contraceptive vaccines based dn cloned zona pellucida genes and the strategy of alloimmunization with zona pellucida polypeptides. In particular, the present invention relates to a contraceptive vaccine for use in a mammalian female comprising a polypeptide which displays at least one epitope for binding of an antibody that inhibits fertilization of an oocyte by a sperm. This epitope is from a zona pellucida protein of the species in which the said vaccine is used.
This invention relates, more particularly, to such vaccines wherein the zona pellucida protein is either the mouse ZP2 protein, the mouse ZP3 protein, the human ZP2, the human ZP3 protein, or homologues of these proteins found in other mammalian species. Further, this invention includes vaccines comprising a synthetic peptide that displays an epitope for such an antibody that inhibits fertilization. In addition, this invention relates to cloned DNA segments variously encoding the mouse ZP3 or ZP2 proteins, or the human ZP2 or ZP3 proteins.
2. Background Information
There is currently much interest in the development of a safe and effective contraceptive vaccine for control of diverse mammalian populations. Contraceptive vaccines would be useful under certain circumstances where relatively long-term but not permanent contraception is desired without the need for frequent intervention, for example, in pets including cats and dogs, in agriculturally important livestock such as cattle and pigs, and in human beings. A contraceptive vaccine preferably should have an effect which is long-lasting and highly specific. Further, to minimize possibilities for birth defects in the event of failed contraception, the antigen which is selected as the immunogen should produce contraceptive antibodies that inhibit fertilization of the egg by a sperm rather than by an abortifacient mechanism involving disruption of early development. In addition, the vaccine preferably should induce an immunological response that is sufficient to be effective for contraception without eliciting a cytotoxic response that might result in abnormal reproductive function.
The mammalian zona pellucida, which surrounds growing oocytes and ovulated eggs, has been recognized as a potential immunogen for a contraceptive vaccine (C. J. Henderson, et al., J. Reprod. Fert. 83: 325-343 (1988); B. S. Dunbar, 1983, Mechanisms and Control of Animal Fertilization, J. F. Hartmann, ed., pp. 140-175, Academic Press, New York; A. T. Sacco, Am. J. Reprod. Immunol. Microbiol. 15: 122 (1987); Millar et al., Targeting of zona pellucida for immunocontraception, in Immunology of Reproduction, Naz, R. K. (ed.), pp. 293-313 (1993)). At birth the mouse ovary contains 10,000-15,000 oocytes in the prophase of the first meiotic division. As cohorts (10-15) of these oocytes enter into a two week growth phase, they synthesize and secrete zona proteins to form the extra-cellular zona pellucida which ultimately reaches a thickness of 7 .mu.m in the fully grown oocyte. The zona is unique to the ovary, being highly antigenic and accessible to circulating antibody during the two week intra-ovarian oocyte growth phase prior to meiotic maturation and ovulation.
Passive immunization of mice or hamsters with anti-zona sera has been shown to produce reversible contraception without obvious side effects. For example, U.S. Pat. No. 3,992,520 to Gwatkin discloses, inter alia, an anti-serum composition for short-term control of fertility comprising antibody obtained by immunizing an animal with water solubilized zona pellucida of a distinct donor species. This method requires isolation of large amounts of a relatively scarce natural antigen which would not be feasible for certain mammals such as humans. Further, long-term administration of antibodies from a foreign (i.e., "heterologous") species leads to induction of reactive antibodies that will inhibit the contraceptive action of the contraceptive antibodies. Further, administration of serum or products isolated from serum carries inherent risks of transmission of blood-born diseases.
Structural information about the zona pellucida has been available for some years. The mouse zona, for instance, is composed of three sulfated glycoproteins, designated ZP1, ZP2 and ZP3, (J. D. Bleil et al., Dev. Biol. 76:185 (1980); S. Shimizu et al., J. Biol. Chem. 258:5858 (1983)) which play important roles in fertilization and early development and have average M.sub.r s of 200,000, 140,000, and 85,000, respectively. ZP2 and ZP3 appear to be complexed into long filaments which are cross-linked by ZP1 in the zona matrix providing structural integrity to the zona pellucida. Sperm initially bind to ZP3 via O-linked oligosaccharide chains and continued binding involves ZP2 as a secondary sperm receptor. Subsequently, ZP3 induces lysis of the sperm's acrosome which releases enzymes (such as glycosidases and proteases) which are thought to be important for the penetration of the zona pellucida by sperm. Following fertilization, both ZP2 and ZP3 are biochemically modified to prevent additional sperm binding and thereby to facilitate the post-fertilization block to polyspermy.
The zona pellucida in other mammals besides the mouse is known to comprise several distinct glycoproteins components with apparent sizes and, hence naming terminologies, that do not necessarily correspond directly to the mouse ZP1 (185-200 kDa), ZP2 (120-140 kDa) and ZP3 (83 kDa) proteins. The human zona pellucida is composed of three proteins designated ZP1 (90-110 kDa), ZP2 (64-76 kDa) and ZP3 (57-73 kDa)(Shabanowitz et al., J. Reprod. Fertil. 82:151-61 (1988); Shabanowitz, 43:260-70 (1990)) and other species in which zona proteins have been characterized include hamster (Moller et al., 137:276-86 (1990), pig (Dunbar et al., Biol. Reprod. (1981); Hedrick et al., Dev. Biol. 121:478-88 (1987); Yurewicz et al., J. Biol. Chem. 262:564-71 (1987), rabbit (Dunbar et al., Biol. Reprod. 24:1111-24 (1981) and horse (Millar et al., J. Reprod. Fert. 96:815-25 (1992)). The correspondence of specific zona proteins among different species is becoming clearer as additional information on the primary amino acid sequence is deduced from cloned zona pellucida genes (Ringuette et al., Proc. Natl. Acad. Sci. U.S.A. 83:4341-45 (1986); Ringuette et al., Dev. Biol. 127:287-95 (1988); Chamberlin et al., Proc. Natl. Acad. Sci. U.S.A. 87:6014-18 (1990); Chamberlin et al., Dev. Biol. 131:207-14 (1989); Liang et al., Mol. Cell. Biol. 10:1507-15 (1990); Liang et al., Dev. Biol. 156:399-408 (1993); Kinloch et al., Dev. Biol. 142:414-21 (1988); Schwoebel et al., J. Biol. Chem. 266:7214-19 (1991); Kinloch et al., Dev. Biol. 142:414-21 (1990)) and direct sequencing of peptides derived from zona pellucida proteins (Ringuette et al., supra (1986); Yurewicz et al., Mol. Reprod. Dev. 33:182-88 (1992)).
In light of the identification of the distinct murine zona pellucida polypeptides, ZP1, ZP2 and ZP3, further experiments on passive immunization with contraceptive antibodies have been conducted. Specifically, rat anti-mouse ZP2 and anti-mouse ZP3 monoclonal antibodies were injected into female mice and were found to bind specifically to the zonae surrounding growing, intra-ovarian oocytes. After ovulation, the binding of the antibody to the zona persisted; and the presence of these antibodies precluded fertilization by preventing sperm from penetration of the zona pellucida. This contraceptive effect was long-term, lasting approximately 15 mouse estrus cycles, but was eventually reversible. There was no evidence of any adverse effect on the development of fertilized embryos to term and no evidence of abnormal ovarian histology or function. However, the antibody binding sites (i.e., "epitopes") recognized on mouse ZP2 and ZP3 by five different rat anti-mouse monoclonal antibodies that were tested are not present on other mammalian zonae pellucidae (East et al., J. Cell Biol. 98:795-800 (1984); East et al., Dev. Biol. 104:49-56 (1984); and East et al., Dev. Biol. 109: 268-73 (1985)). This species specificity limits the usefulness of these particular antibodies as contraceptive agents essentially to murine species. In addition, even if analogous murine anti-ZP2 or anti-ZP3 antibodies that inhibit fertilization could be identified for ZP2 or ZP3 of non-murine species, there are inherent side-effects from the repeated administration of heterologous antibodies, as noted above.
There have been several studies on active immunization using preparations of isolated zona pellucidae to immunize rodents, rabbits, and primates (C. J. Henderson, et al., J. Reprod. Fert. 83:325 (1988); R. B. L. Gwatkin, et al., 1977, Fert. Steril. 28:871 (1977); Drell et al., Biol. Reprod. 30:435-44 (1984); Sacco et al., Biol. Reprod. 36:481-90 (1987); Jones et al., J. Reprod. Fertil. 95:513-25 (1992)).
Further, the U.S. Patent to Gwatkin cited above (U.S. Pat. No. 3,992,520) also discloses a vaccine for the immunological control of fertility in female mammals that consists of an aqueous solution of water solubilized zona pellucida prepared by heating mammalian zone pellucida at 65-100.degree. C. in an aqueous medium. One example therein describes a bovine antigen preparation intended for use in humans.
U.S. Pat. No. 4,996,297 of Dunbar is limited to three rabbit cDNA sequences S1, P2, and P3 thought to encode rabbit zona proteins, to the use of these cDNAs to produce polypeptides that contain epitopes on three rabbit zona proteins (50 kDa, 75 kDa, and 80 kDa), and to the use of the recombinant polypeptides to vaccinate other mammals in order to elicit antibodies that bind to that mammal's zona pellucida for contraception (i.e., heteroimmunization).
Japanese Patent 63,150,299 discloses a pig zona pellucida antigen for use as a contraceptive vaccine for pigs or humans that is characterized as a glycoprotein of 20 to 30 kDa in molecular weight which can be extracted from soluble pig zona pellucida with 8.5M urea and 2% 2-mercaptoethanol.
Despite positive results under experimental conditions, methods of preparing a vaccine from natural zona pellucida materials are clearly difficult if not outright impractical for commercial use, particularly in the human case, due to limited sources of antigen and to difficulties in quality control of such poorly defined vaccines. Further, wide-spread ovarian histopathology and dysfunction were reported in rabbits, dogs and primates after active immunization with zonae pellucidae or extracted antigens (see, for example, R. B. L. Gwatkin, et al., Gamete Res. 1:19 (1980); A. T. Sacco, Am. J. Reprod. Immunol. Microbiol. 15:122 (1977)). Several studies have suggested that both the dose and the purity of the immunogen contributed to these abnormalities, two properties that are particularly difficult to control in such relatively crude antigen preparations.
The effect of the genetic origin of the zona pellucida antigen on its ability to immunize a given species against conception has been examined in several studies. For instance, the efficacies of contraceptive immunizations with pig and rabbit zonae pellucidae on fertility in rabbits was compared. This comparison of results with "alloimmunization" (literally "self-immunization", using antigen from the same species, i.e., an "alloantigen") with those of "heteroimmunization" (using antigen from another species, i.e., an "heterologous" antigen) suggested (D. M. Wood et al., Biol. Reprod. 25:439-450 (1981)) that heteroimmunization of rabbits with porcine zonae is more effective in reducing fertility than alloimmunization with rabbit zonae. More recent work using immunoaffinity purified antibodies to zona pellucida to compare immune responses in alloimmunization of male and female rabbits has continued to support the greater effectiveness for contraception of heteroimmunization with zona pellucida antigens. (S. M. Skinner, et al., J. Reproductive Immunology 12:81-92 (1987)).
Another general approach toward providing a vaccine related to any antigen involves the use of a particular type of antibody, called an "anti-idiotypic" antibody, as an immunogen to actively immunize an animal. Anti-idiotypic antibodies are antibodies directed to the antigen binding site of another antibody; accordingly, the antigen binding site of the anti-idiotypic antibody mimics or represents an image of the site on the antigen that is bound by the other antibody. U.S. Pat. No. 4,795,634 to Grimes et al. (equivalent of WO 87/05,516) discloses a vaccine that comprises anti-idiotypic antibodies to anti-zona pellucida antibodies to express images of zona pellucida antigens. This vaccine suffers from drawbacks including the fact that anti-idiotypic antibodies are generally difficult and expensive to prepare in amounts and purity satisfactory for vaccine usage, particularly in human applications. Further, heteroimmunization with antigens comprising antibodies from another species may induce predominantly antibodies to sites on the antibody other than the desired target, the antigen binding site. In other words, the desired antigen binding site may not constitute an "immunodominant" antigenic site (or "determinant") for the vaccine antibody protein in a species different from that which produced the vaccine protein (see below for a discussion on the basis of immunodominance). (See also U.S. Pat. No. 4,996,297 of Dunbar et al.)
Another technique for producing vaccines that is known generally in the art is the use of specific isolated polypeptides as antigens, or of peptides representing portions of such polypeptides, in place of crude antigen preparations comprising aqueous extracts of target tissues. Accordingly, European Patent EP-0117934 to Stevens discloses a modified antigen for use in fertility control comprising an unspecified antigen from the zona pellucida, or a peptide having a sequence corresponding to at least part of the sequence of such a zona pellucida antigen, which antigen or peptide has been chemically modified outside the body of the animal. The modified antigen has a greater capacity to induce antibodies than the unmodified antigen from which it is derived. According to the specification and claims, such modification includes coupling the antigen or peptide through a maleimido linkage to a suitable "carrier" protein that is biologically foreign to the animal to be vaccinated and of size sufficient to elicit antibody response. Neither this European application nor any related applications, as yet published, teaches specific zona pellucida polypeptides or peptides that are suitable for use as contraceptive vaccines.
In light of the complexities, difficulties and uncertainties of all the contraceptive vaccines described above, there is yet a need for a simpler, safer, cheaper, more defined and effective contraceptive vaccine. The present invention is based on the premise that vaccination with a "self" zona protein (alloimmunization) is most likely to elicit antibodies that will cross-react with the native zona pellucida and prevent fertilization. Furthermore, by using relatively short peptides as immunogens, the adverse effects on ovarian structure and functions, at least some of which can result from a T cell mediated autoimmune response, can be avoided. However, the success of this approach depends on knowledge of the primary amino acid sequence of the zona pellucida proteins. Because of the paucity of biological material, this sequence information can only be obtained by cloning cDNAs encoding the zona proteins and deducing the amino acid sequence from the nucleic acid sequence. Toward this end, the present inventor and associates have recently constructed a mouse ovarian cDNA expression library and isolated two overlapping ZP3 cDNA clones (M. J. Ringuette et al., Proc. Natl. Acad. Sci. USA 83:4341 (1986)), one of which expresses a fusion protein recognized by an anti-ZP3 monoclonal antibody (East et al., Dev. Biol. 109: 268 (1985)).
The identity of these clones was confirmed by a comparison of the amino acid sequence encoded by a 60 nucleotide stretch of their nucleic acid sequence with the terminal amino acid sequence (20 amino acids) of a large internal fragment isolated from the ZP3 protein (Ringuette et al., supra 1986)). This fragment was isolated from purified ZP3, following digestion with a protease, by affinity chromatography using an anti-ZP3 monoclonal antibody. Therefore, it was clear that this fragment was capable of expressing an epitope for a contraceptive antibody; however, the location of that epitope within scores of amino acid residues was not known, and as disclosed herein, is distinct from the 20 amino acid sequence obtained. More importantly, the ability of this proteolytic cleavage fragment to serve as an immunogen in a vaccine was not known, nor was there any practical means for preparing sufficient material from natural sources to test that cleavage fragment further.
A first attempt to utilize the cloned mouse ZP3 cDNA described above to produce a vaccine was unsuccessful (S. M. Chamow and J. Dean, 1987, abstract of presentation to the American Society of Biological Chemists). This effort involved testing of the recombinant ZP3-.beta.-galactosidase fusion protein, which contained most of the ZP3 amino acids as well as a larger portion of .beta.-galactosidase and was generated according to well known methods in genetic engineering that have successfully produced other antigens with native immunoreactivity. Immunization with this particular fusion protein, however, failed to induce detectable antibodies that would react with native ZP3; reactivity was detected only after reduction of disulfide bonds and denaturation.
The basis of this failure to induce anti-ZP3 contraceptive antibodies, despite that fact that the cDNA clearly encoded a proteolytic cleavage fragment that reacted with such an antibody, is not entirely clear. It may be that, under the conditions of immunization, the portion of the fusion protein that encoded the contraceptive antibody epitope did not assume the proper conformation to react with such antibodies. In other words, although the fusion protein surely encoded the amino acids that formed the epitope in the native ZP3 protein, it may be that those amino acids did not exhibit (i.e., did not "display") that epitope in this instance. It is also possible that epitopes for other antibodies, which were located on the .beta.-galactosidase moiety of the fusion, may have been immunodominant over the contraceptive antibody epitopes and thus prevented a detectable contraceptive antibody response (see discussion of immunodominance below). Finally, a combination of these effects and others may have united to prevent the desired contraceptive antibody response to the fusion product of the recombinant DNA which expressed most of the ZP3 polypeptide. These results clearly illustrate the unpredictability of the immunogenicity of a polypeptide under any given set of conditions, no matter how efficacious they may be for other antigens, and the need for experimental determination of the necessary physical form of the amino acids that encode an epitope (e.g., polypeptide size and nature of attached amino acid sequences) to display that epitope and, further, to induce antibodies to it.
Accordingly, it is an object of the present invention to find an efficacious way to use contraceptive antibodies and cloned genes encoding zona pellucida proteins to develop contraceptive vaccines for use in a mammalian female. More particularly, it is an object of this invention to provide such vaccines comprising polypeptides that include defined amino acid sequences that are selected for their ability to display epitopes for contraceptive antibodies.
Additional immunological analyses of the individual ZP polypeptide components have been carried out. For example, specific monoclonal and polyclonal antibodies have been employed to define distinct antigens of the porcine zonae pellucidae, leading to the suggestion that there are both unique and shared antigenic determinants present in the individual components of the zona pellucida, but that the immunodominant determinants appear to be unique to each glycoprotein (T. M. Timmons, et al., Biology of Reproduction 36:1275-1287 (1987)).
Finally, there has been a report of an effort to molecularly clone cDNAS encoding specific antigenic sites from rabbit ZP proteins using antibodies that recognize determinants found on ZP antigens of several species (P. Cheung et al., 1987, abstract of a presentation at the twenty-seventh annual meeting of the American Society for Cell Biology, St. Louis, Mo., November 16-20, J. Cell Biol. 105, no. 4 part 2, 334A). This abstract reported in part that:
"These studies demonstrated that cross-species affinity purification of antibodies is an effective method for isolating cDNA clones expressing antigens which are shared among different mammalian species."
However, no specific nucleotide or amino acid sequences were disclosed in this abstract, nor was the contraceptive potential of the antibodies discussed; indeed, there was no mention of any contraceptive vaccine.
In a speculative exposition on the use of recombinant DNA and synthetic peptide technologies for development of a human contraceptive vaccine from porcine zona pellucida antigens (C. J. Henderson, et al., J. Reprod. Fert. 83:325 (1988)), the identification of amino acid sequences displaying epitopes for contraceptive vaccines on a particular porcine polypeptide is anticipated, although absolutely no sequences of the polypeptide are disclosed. Nevertheless, this reference goes on to hypothesize that known vaccine technologies, including synthetic peptides and vaccinia virus expression vectors, will provide successful human vaccines based on this particular porcine polypeptide that is known to be immunologically related to human zona pellucida antigens. Furthermore, while asserting that monoclonal antibodies to this polypeptide that exert a contraceptive effect "will be extremely important in defining the epitopes with contraceptive potential . . .", this report also notes that, despite obtaining monoclonal antibodies reactive with this polypeptide, the authors "have failed to generate a monoclonal antibody with contraceptive effect; this is in accord with other published reports . . . ."
Although a complete exposition of the current theoretical basis of immunogenicity and antigenicity of polypeptides is beyond the scope of the present disclosure, a brief discussion of selected principles and terms of this active art will facilitate further understanding of the instant invention. In this application, absent an express statement to the contrary, each use of the term "polypeptide" encompasses any polymer comprising two or more amino acids coupled by peptide linkages (i.e., dipeptides, oligopeptides, peptides, polypeptides) as well as proteins consisting of multiple polypeptide subunits.!
Accordingly, it should be noted, first, that the necessary and sufficient properties of a polypeptide for inducing antibodies cannot be predicted for any given set of conditions (e.g., for a particular species, or for presentation in a certain form). Nevertheless, much more has been learned about this subject in the past decade than is reflected in any of the art cited so far herein, and it is a further object of the present invention to exploit aspects of this knowledge for design of advantageous contraceptive vaccines.
In particular, comprehension of the present invention will be aided by the now widely held view that the nature and level of the immune response to a polypeptide depends on its interactions with at least two distinct classes of immune system cells, namely B-cells and T-cells. In simple terms, the role of B-cells in immunity may be thought of as recognition of the specific sites on macromolecules to which antibodies are produced and subsequent production of those antibodies. These B-cell recognition sites, which provide the main basis for immune recognition of non-self molecules and are also called B-cell epitopes, are of a size corresponding to about that of the antigen binding site on an antibody, typically of a diameter equivalent to the length of a peptide containing about four to six amino acids.
It may be noted here that there exists a formal distinction between the epitope for a B-cell and that of its related antibody. In other words, due to complex biological mechanisms that intervene between the recognition by a B-cell of a given site on an antigen and the consequent production of antibodies to that site, it is possible that the ultimate antibody recognition site may not be precisely identical to the initially recognized B-cell epitope. However, for the present purposes, a B-cell epitope may be considered to be essentially the same structure as the binding site for the corresponding antibody.!
The functions of T-cells, on the other hand, relate in large measure to helping to activate antibody production by B-cells upon initial exposure to an antigen, as well as to enhancing their antibody response upon subsequent reexposures (i.e., to "immune memory" or the "amnestic" response). To play their roles in immunity, T-cells must also recognize specific sites on an antigen to which antibodies are produced, and such T-cell epitopes are about the same size as B-cell epitopes.
B-cell and T-cell epitopes on any given polypeptide, however, need not comprise the same amino acid residues. In fact, it will be appreciated by those of ordinary knowledge in the current art of peptide immunology at the molecular level, that even in a peptide consisting of only half a dozen amino acids, there may coexist several different B-cell epitopes (comprising, for instance, from two to four atoms that contact complementary structures on the antibody) and one or more distinct T-cell epitopes which may or may not include atoms of amino acids also included in a B-cell epitope.
It is also well known that the vast majority of small peptides (containing six to twenty amino acids, for instance) that have been tested for induction of antibodies are considerably less potent immunogens than the larger proteins from which they have been derived, despite ample ability of the peptides to bind to antibodies directed against those larger proteins. Certain chemical modifications of a peptide, particularly coupling of the peptide to a larger proteinaceous "carrier", generally enhance the immune response to a small peptide.
Although the role of such a carrier still may not be fully understood in all respects, it has been clearly established, in particular, that there is no specific minimum size requirement for peptides in general to induce a substantial immune response. Rather, it is now widely believed that a major function of the carrier is to provide T-cell epitopes in close association with the B-cell epitopes on the short peptide which is statistically unlikely to contain both T-cell and B-cell sites recognized by the immune system of any given individual.
It may also be noted here that it has been shown that a T-cell epitope taken from one protein, in the form of a short peptide, may be combined with a short peptide comprising a B-cell epitope of another protein, to form a single peptide that induces a more complete and higher level immune response than either peptide alone.
More broadly, it is now widely accepted that the capability of any individual to mount any immune response to a given epitope, as defined by a precise configuration of a small number of atoms, depends ultimately on the genetic make-up of the immune system genes which separately control the specificities of antigen recognition by B-cells and T-cells. Further, it is understood that the ability of a given B-cell epitope to induce cognate antibodies (i.e., antibodies which recognize that epitope) also depends upon the context within which that epitope is presented to the immune system, in terms of both associated T-cell epitopes and other B-cell epitopes. The latter sites may be "immunodominant" relative to the selected B-cell epitope of interest, that is, they may contend more effectively for the attention of the immune system than the selected B-cell epitope and thereby distract limited system resources from mounting the desired response to that selected epitope. In other words, B-cell epitopes that do not induce detectable antibodies in the presence of other, so-called immunodominant epitopes, which frequently occur in large polypeptides, often do induce significant levels of cognate antibodies when presented in a different context that lacks such immunodominant sites, on a short peptide, for example.
In conclusion, it is a further object of the present invention to exploit various consequences of the above noted characteristics of and distinctions between B-cell and T-cell epitopes, as well as methods for predicting and actually detecting amino acid sequences that serve as T-cell or B-cell epitopes. These will be discussed further below, as needed, in relation to the description of the present invention.